Direct answer: First, diagnose where patients are actually being lost, because “we need more patients” is almost never the real problem. Pull three numbers this week: time-to-first-contact on referrals, referral-to-screen conversion by site, and screen-fail reasons. Then triage your sites, fix follow-through before buying more advertising, and set a hard date: if the enrollment curve hasn’t bent within eight weeks, bring in outside help, because every month behind costs a sponsor far more than any rescue engagement will.

You’re in the majority, for what it’s worth

Nearly 85% of interventional trials miss their enrollment milestones. That’s not a comfort, it’s a warning about how this usually goes: months of “next month will be better,” a mid-study ad-budget increase that produces referrals nobody converts, then a timeline extension that costs more than the entire recruitment budget would have. The trials that recover are the ones that treat a slipping curve as an emergency in month two instead of month eight.

Week one: find the real leak

Get these three numbers before making any decision:

1. Time-to-first-contact. For your last 50 referrals, how many hours passed before a human called them? If the median is over 24 hours, this is your problem, and no amount of new spend fixes it. For calibration: our average from form submission to first phone call is 87 seconds. You don’t need to hit that. But the distance between 87 seconds and four days is the distance between enrolling and apologizing.

2. Conversion by site. Referral-to-screen and screen-to-enrollment, per site, not averaged. In almost every behind-schedule trial we’ve rescued, a minority of sites produce nearly everything while several produce zero. Averages hide this.

3. Screen-fail reasons, top three. If patients fail on criteria a phone pre-screen could catch, your funnel is burning site capacity on people who were never eligible.

Weeks two to four: triage

Fix follow-through first. Same-day callbacks on every referral. One named owner per patient from click to clinic. If your current vendor’s answer to “who calls the patient back” is “the site,” you’ve found the leak.

Sort your sites into three buckets. Producers (protect and feed them), fixable (a specific blocker like coordinator bandwidth or scheduling), and dead (stop routing patients there, and consider whether the enrollment plan needs a site the plan doesn’t have).

Pre-qualify harder, not softer. Counterintuitive when you’re behind, but flooding sites with marginal referrals slows enrollment further. Double-screening cut screen fails by 18% in our programs, and 89% of the patients we send to a site go on to sign the ICF. That’s what the site hours you have left should be spent on: patients who can actually randomize.

Re-cut the creative. If your ads read like protocol summaries, replace them with patient-story creative. Same budget, different response.

Week eight: the honest checkpoint

If the weekly enrollment rate hasn’t visibly bent by now, more of the same won’t bend it. Your options:

Amend the protocol or add sites. Sometimes the criteria are the problem. This is slow and expensive but occasionally it’s the truth.

Bring in a rescue specialist. This is most of what we do at Clinical Enrollment, and it’s why our model is built the way it is: we take ownership of the whole path, patient identification through site follow-through, and we only get paid for enrolled patients. A sponsor who’s already been burned by referral-count invoices shouldn’t have to take vendor risk twice. On programs where we start at launch we typically deliver 32% of a trial’s total enrollment, and as much as 47%; rescue is the same machine pointed at a tighter window. Across rescue and standard programs, our sponsors have averaged $19.2M in savings versus their projected overrun, with enrollment finishing 57 days ahead of the revised plan.

Extend the timeline. The default outcome if nothing changes. Price it honestly: site fees, CRO burn, staff time, and the cost of your drug reaching patients later. That number makes every other option look cheap.

What to demand from any rescue vendor

• Payment tied to enrollments, not referrals or ad impressions
• A named owner for every patient handoff
• Site-level conversion reporting, weekly
• A forecast they’ll put their fee behind
• References from sponsors whose trials were behind when they arrived

If a vendor won’t tie their fee to the outcome, they’re forecasting with your money.

Frequently asked questions